The Endless Baby Boomer Generation

ABSTRACT

Using the Eurobarometer cross-sectional data sets conducted between 1976 and 2008 in nine European countries, we aim to describe and explain cohort differences. We make use of recent improvements of the age–period–cohort (APC) methodology to have a better assessment of these cohort-based changes. At first, we find strong and significant cohort fluctuations in participation in political discussion where the people born in the period 1945–1955 have an almost systematically higher participation than previous and following cohorts. The introduction of individual level variables such as education does not explain the cohort bump of the early baby boomers. Post-APC estimations show that the contextual factors size of a birth cohort and economic affluence at the entry into adulthood are important explanatory factors for cohort differences. In conclusion, since strong cohort effects are detected in the APC model, the decline in participation of the young appears to be less a temporary moratorium that will vanish with age but a durable trait without redeem.

Key words:

• cohort analysis
• socialization
• demography
• education
• political participation

Notes

1. The authors mention a ‘moratoire politique’, a moratorium in politics that derives from Erik Erikson’s (1950, 1956) development stage theory of psychosocial moratorium of the young.

2. Additional analyses in which the frequently discussers are contrasted to the occasionally and never discussers show similar results. We estimated the same models with a continuous dependent variable as well. The best solution would be to make use of an ordinal logistic regression specification, but due to the limitations in the general linear model with constrains in Stata we did not have this opportunity.

3. The APCD is available as a Stata ado-file. It can be downloaded by typing ‘ssc install apcd’ in Stata.

4. The constraint Slope_a(α_a) = 0 means the trend of the age effect is zero and is true only if . This constraint is easily expressed as a linear equation of coefficients.

5. This ‘synthetic cohort’ tool is a common descriptive method in demography, sociology and epidemiology (Mason and Fienberg 1985; Preston et al. 2001). The horizontal axis represents age and the vertical one a dependent variable (such as intensity of political participation). Curves represent the trajectory of birth cohort groups, so we can observe the differences in aging process. The cohort diagram is an alternative where cohort is on the horizontal axis, and the curves present age groups, so we can compare different cohorts when they have the same age.

6. The height of the variance inflation factor (1.19) shows that the model does not suffer from multicollinearity.

Additional information

Funding

This work was supported by the Luxembourg Fonds National de la Recherche (FNR) [FNR/P11/05 and FNR/P11/05bis].

Notes on contributors

Louis Chauvel

Louis Chauvel PrDr is Professor at the University of Luxembourg and head of the Institute of Research on Socio-Economic Inequalities (IRSEI), did his PhD and Habilitation in the University of Lille and in Sciences Po Paris. He is an honorary member of the French Institut universitairede France. He is a specialist of social inequality, birth cohort dynamics and welfare regime sustainability.

Fransje Smits

Fransje Smits is Postdoctoral Researcher in the University of Luxembourg, Institute of Research on Socio-Economic Inequalities (IRSEI). She is a specialist of politico-religious transformations in the Netherlands and of depression issues in postmodern societies.